EARLY DETERMINATION OF STROKE SUBTYPE: CT ANGIOGRAPHY

University Of Pennsylvania, Philadelphia PA

Investigators

Linked publications & trials

Abstract

The candidate is a stroke fellowship-trained neurologist dedicated to a career in acute stroke diagnosis and treatment. This application proposes a comprehensive program designed to evolve the candidate into an independent investigator in patient- oriented stroke research. The University of Pennsylvania has outstanding clinical, radiological, and epidemiological resources to support this proposal. The training component of this plan entails formal graduate education in clinical epidemiology and research design, specifically focusing on vascular disease and technology assessment. The research component is a prospective evaluation of a new diagnostic tool in the context of stroke management. The central hypothesis of the research proposal is that early clinical determination of stroke mechanism is limited, but computed tomography angiography (CTA) will effectively identify patients with specific stroke mechanisms for specific mechanism- directed therapies. Ischemic stroke may be caused by one of several different etiologies, and treatment may be best directed toward the underlying mechanism. Therapy for acute stroke must be initiated in the first few hours after stroke onset, but current methods for identifying stroke mechanism are inaccurate during this critical time period. The gold standard method for determination of stroke mechanism requires a thorough evaluation that cannot be obtained within the therapeutic time window. New imaging techniques may be useful for the early determination of stroke subtype, but there has been no formal analysis of their utility in stroke triage and treatment. An extremely promising novel technique is CTA, a quick noninvasive method that allows for visualization of the cerebral vasculature. Specific Aims: The role of CTA in the early diagnosis of ischemic stroke subtype will be determined by measuring its positive predictive value (primary outcome measure), sensitivity, specificity, and reliability. We hypothesize that these characteristics are superior for CTA than current methods of clinical stroke diagnosis. CTA-based diagnosis of stroke mechanism will be compared with early clinical diagnosis, and ultimately will be compared with the gold standard final diagnosis. A cohort of 150 patients will be examined with CTA within 24 hours of the onset of stroke. Investigators will make an early assessment of stroke mechanism based on the available clinical information and will be blinded to CTA results. Stroke mechanism will ultimately be established after a diagnostic evaluation is completed. Investigators will then use both clinical and CTA data to determine a CTA-based diagnosis. The positive predictive value (PPV) of the clinical and CTA-based diagnoses will be compared, adjusting for multiple raters and multiple patients. This approach will, for the first time, formally evaluate CTA as a tool for early stroke diagnosis and selection of patients with specific stroke mechanisms for specific mechanism-directed stroke therapies.